Journal of the Korean Operations Research and Management Science Society (한국경영과학회지)

The Korean Operations Research and Management Science Society (한국경영과학회)
권호 표지

A Psychometric Method for Structuring Expert Knowledge:Application to Developing Credit Analysis Espert System for Small-Medium Companies Using Nonfinancial Statement Information

계량심리학의 방법론을 이용한 체계적인 전문가 지식구조분석 방법 : 비재무항목을 활용한 중소기업 신용평가전문가시스템 규칙개발에 적용

  • 이훈영 (경희대학교 정경대학 경영학부) ;
  • 조옥래 (조흥은행 인력개발부 연구원) ;
  • 이시환 (경희대학교 정경대학 회계학과)
  • Published : 1998.03.01
Download PDF
⟨ Previous Next ⟩

Abstract

Translating expert knowledge into production rules has been the most difficult and time-consuming when building expert systems (Buchanan et al. 1983). Especially, buidling hierarchical structure, i. e. developing sequential or dominant relationship among production rules is one of the most important and difficult processes. Hierarchical relationship among rules has been typically determined in the course of interviewing human experts. Since this interviewing procedure is rather subjective, however, the hierarchically structured rules produced in terms of interviewing is widely exposed to the severe discussion about their validity (Nisbett and Wilson 1977 : Ericsson and Simon 1980 : Kellog 1982). We thus need an objective method to effectively translate human expert knowledge into structured rules. As such a method, this paper suggests the order anlaysis technique that has been studied in psychometries (Cliff 1977 : Reynolds 1981 : Wise 1983). In this paper we briefly introduce the order analysis and explain how it can be applied to building hierarchical structure of production rules. We also illustrate how bankrupcy prediction rules of small-medium companies can be developed using this order analysis technique. Further, we validata the effectiveness of these rules developed by the order analysis, in comparison with those built by other methods. The rules developed by the proposed outperform those of the other traditional methods in effectively screening the bankrupted firms.

Keywords

References

  1. 전문가시스템의 원리와 개발 이재규(외5인)
  2. Building Expert Systems Constructing an Expert System Buchanan, B. G.;Barstow, D.;Bechtal, R.;Bennet, J.;Clancy, W.;Kulikowski, C.;Mitchell, T.;Waterman, D. A.;Hayes-Roth, F.(ed.);Waterman, D. A.(ed.);Lenat, D. B.(ed.)
  3. Artificial Intelligence and Statistics Use of Psychometric Tools for Knowledge Acquisition: A Case Study Butler, A. K.;Corter, E. J.;Gale, William(ed.)
  4. Reliability and Validity Assessment Carmines, G. Edward;Zeller, A. R.
  5. AAA-84: Proceedings of the National Conference on Artificial Intelligence Claasification Problem Solving Clancy, W. J.
  6. Psychometrika v.2 A Basic Theory Generalizable to Tailored Testing Cliff, N.
  7. Mathematical Psychology: An Elementary Introduction Coombs, C. H.;Dawes, R. M.;Tversky, A.
  8. Psychometrika v.51 no.3 Extended Similarity Trees Corter, J.;Tversky, A.
  9. Marketing Science v.11 no.2 A New Multidimensional Scaling Methodology for the Analysis of Asymmetric Proximity Data in Marketing Research DeSarbo, W.;Manrai, A.
  10. Psytchological Review v.87 Verbal Reports as Data Ericsson, K. A.;Simon, H. A.
  11. American Sociological Review v.9 A Basis for Scaling Qualitative Data Guttman, L.A.
  12. Journal of Mathematical Psychology v.20 Monotonic Models for Asymmetric Proximities Holeman E.W.
  13. Memory and Cognitition v.10 When Can We Introspect Accurately About Mental Processes? Kellog, R. T.
  14. Psychological Review v.84 Telling More Than We Can Know: Verbal Reports fo Mental Processes Nisbett, R. E.;Wilson, T. D.
  15. Educational and Psychological Measurement v.41 ERGO: A New Approach to Multidimensional Item Analysis Reynolds, T. J.
  16. SAS Language Reference Version 6(First Edition) SAS Institute Inc.
  17. Psychometrika no.42 Additive Similarity Trees Sattath, S.;Tversky, A.
  18. Applied Psychological Measurement v.7 no.3 Comparisons of Order Analysis and Factor Analysis in Assessing the Dimensionality of Binary Data Wise, S.